International Conference on Materials for Advanced Technologies 2011, Symposium O
Tin Sulphide Thin Films Synthesised using a Two Step Process
M. Leacha, K.T. Ramakrishna Reddyb,c,*, M.V. Reddyb, J.K. Tanc, D.Y. Janga and R.W. Milesc
a Seoul National University of Science and Technology,172 Gongneung-dong 2, Nuwon-gu, Seoul, South Korea b Department of Physics, Sri Venkateswara University, Tirupati 517502, India
c School of CEIS, Northumbria University, Newcastle, NE1 8ST, United Kingdom
Over the past few decades, CdTe and CuInGaSe2 (CIGS), deposited in the form of polycrystalline thin films, have been developed for use as absorber layer materials in photovoltaic solar cell devices. This has resulted in the production of commercial modules with efficiencies > 10 % for both technologies [1]. In
372 M. Leach et al. / Energy Procedia 15 (2012) 371 – 3782 M. Leach et al. / Energy Procedia 00 (2011) 000–000
the case of CdTe, the thin-film modules are now produced more cheaply than silicon-based modules. There remains however concerns in some countries as to the desirability of using cadmium, a heavy metal, in these devices. In the case of CIGS there are concerns with respect to the lack of availability of Ga and In, as this will limit the large-scale use of this product in the future. These drawbacks have led to research groups across the world investigating the potential of new materials that consist of elements that are both abundant and environmentally acceptable for use in such devices. One of the many possible materials is tin sulphide, SnS [2]. Like CdTe and CIGS, SnS has a near-optimum direct energy bandgap (1.35 eV), it is amphoteric, and it can be deposited by a wide range of chemical and physical deposition methods. To date the most commonly used methods for depositing SnS are chemical bath deposition, electrodeposition, spray pyrolysis and thermal evaporation [3-7]. In all these cases the photovoltaic effect has been observed and devices with efficiencies up to 2% produced using some of these methods [2]. In this paper we have investigated the use of a less well investigated two step process for making SnS, based on the use of sulphidising thin films of tin that have been pre-deposited onto glass and Mo-coated glass substrates. In particular, the formation of SnS by annealing pre-deposited tin layers in a mixture of hydrogen sulphide (H2S) and argon (Ar) gas environment was investigated. The physical and chemical properties of the resulting films have also been compared with those obtained by annealing the tin layers in elemental sulphur. These routes for producing tin sulphide are compatible with existing industrial processes, as both the thin film deposition of tin layers and the sulphidisation processes are well established for other applications.
2. Experimental details
The tin layers used in the present study were deposited by DC magnetron sputtering using a SORONA SRN-110 system. The Sn target was 99.99 % pure, it was 18 cm in diameter and 2 mm thick. Mo-coated soda lime (SL) glass slides were used as substrates. The layers of Sn were deposited at a base pressure of 810-6 Torr using argon as the sputtering gas. The sputtering was carried out at an argon pressure of 5 mTorr and with a sputtering power of 1500 W. The tin layer on glass (measuring 25 mm 25 mm) was placed in a customised vacuum tube furnace (Sun Han Vacuum Tech Ltd) for sulphidisation. The quartz furnace was initially evacuated to 0.5 Torr and then flushed with Ar gas. The sulphidisation was carried out for temperatures in the range 300-450 °C for a fixed time period of 2 hours. The furnace was heated to the pre-determined temperature, allowed to stabilise, and then flooded with a gas mixture of 5 % H2S in Ar. The gas mixture flow was cut off after the annealing time and the quartz tube flushed with pure Ar gas. Finally the furnace was allowed to cool to room temperature under steady vacuum, prior to the sample being removed. Structural studies (i.e., the phases present, the orientation of the planes and crystal structure of each phase) were investigated using X-ray diffraction (XRD). The XRD measurements were taken using Rigaku Denki Co. D/max-IIIC X-ray diffractometer using CuKα radiation ( = 1.5418 Å). The surface topology and topography of the samples were observed using a JEOL JSM-6700F field emission scanning electron microscope (FE-SEM). An Oxford Instruments energy dispersive X-ray analysis (EDAX) facility attached to the FE-SEM was used to determine the elements present and their concentration by mass in the films. The optical reflectance measurements were made in the wavelength range 500-1000 nm using a Perkin Elmer Lambda-35 UV-Vis spectrophotometer. This data was used to determine the energy bandgap of each film. 3. Results and discussion
The visual observation of Sn-precursor layers deposited on Mo-coated glass substrates by DC magnetron sputtering indicated that the layers were pin-hole free, uniform and strongly adhered (scratch tape test) to the substrate surface compared to the Sn layers grown directly on soda-lime glass substrates.
373M. Leach et al. / Energy Procedia 15 (2012) 371 – 378 M. Leach et al. / Energy Procedia 00 (2011) 000–000 3
Figure 1 shows the XRD patterns of SnS films synthesised at different sulphidisation temperatures. From the XRD spectra it is observed that all the layers had an intense (110) peak. This peak corresponds to the presence of the Mo layer, that had been pre-deposited onto the glass substrates. A strong peak related to the (111) plane of SnS phase was observed in all the sulphidised layers. Peaks corresponding to the (200), (021), (131) and (151) planes were also observed in the layers sulphidised at temperatures of 350 and 400 °C. This despite some of the peaks being masked by the large (110) peak at 40.32 degrees due to the Mo layer. Further, planes corresponding to the (002) and (001) orientations were also observed in the layers synthesised at 300 and 350 oC, respectively. These were related to diffraction from the crystal planes of Sn2S3 and SnS2. Tin layers sulphidised at 400 C had a strong (200) peak of SnS that was not observed at other temperatures. Further, the films formed at Ts = 400 oC had peaks corresponding only to the SnS phase. These had a strong (111) peak at a diffraction angle (2θ) of 31.7 degrees. The films synthesised for Ts ≤ 350 oC had peaks associated with the presence of the Sn2S3 and SnS2 phases although these dissociated gradually when TS approached 400 oC. The observed changes in the formation of the various phases, with the increase of TS, were mainly due to the changes in the composition of the film, particularly sulphur incorporation into the film. The appearance of different phases at different growth temperatures depends on the energies of formation of these phases, which in turn depend on the thermo-dynamic data as reported in the literature [8, 9] on bulk materials. However, a detailed investigation of such thermodynamic data measurement is necessary to correctly predict the growth mechanism for SnS films, particularly in sulphidisation process. At lower sulphidisation temperatures, the presence of additional phases, Sn2S3 and SnS2, in conjunction with the SnS phase might be due to availability of more tin and low energy formation of these phases. However with the increase of TS, more Sn is converted into SnS. A similar behaviour of the formation of different phases with the change of substrate temperature was reported in literature for SnS films grown by different techniques [10-12]. The sudden decrease of (111) peak intensity with the disappearance of other planes in the films obtained at TS = 450 oC was due to the re-evaporation of S from the film surface owing to its high vapour pressure at such temperatures. It was even possible that Sn also had re-evaporated out of the substrate surface, leaving only Mo on the glass, as these films didn’t show any Sn peak.
In parallel to this work, Sn layers deposited on Mo-coated glass by DC magnetron sputtering, were also sulphidised using elemental S. Figure 2 shows the X-ray diffraction spectra of SnS films synthesised in the presence of elemental S using various sulphidisation temperatures. All the spectra showed the (111) peak. The intensity of this peak increased continuously with an increase of the sulphidisation temperature. Tin layer sulphidised at 350 C had only the (111) peak, indicating that most of the crystallites are oriented in the <111> direction. Further, the spectra indicated that Sn layers could be converted into SnS at lower sulphidisation temperatures when using elemental S rather than when using H2S and Ar environment (350 C rather than 400 C). The precursor layers sulphidised at 250 oC had Sn, S and Sn2S3 peaks in addition to those due to the presence of SnS. With the increase of sulphidisation temperatures the Sn and S had reacted together to form the sulphides while simultaneously Sn2S3 dissociated into SnS and S, resulting in single phase SnS. Similar results on the sulphidisation of Sn at lower temperatures were reported in the literature [10, 11].
374 M. Leach et al. / Energy Procedia 15 (2012) 371 – 3784 M. Leach et al. / Energy Procedia 00 (2011) 000–000
Fig. 1. X-ray diffraction pattern of SnS films sulphidised using H2S+Ar at various temperatures.
Fig. 2. XRD spectra of SnS layers sulfurised using elemental S at various temperatures.
375M. Leach et al. / Energy Procedia 15 (2012) 371 – 378 M. Leach et al. / Energy Procedia 00 (2011) 000–000 5
The size of the crystallites (L) in SnS films formed by H2S+Ar gas sulphidisation was evaluated using the Debye-Scherrer relation [12]:
L = (0.94λ)/(β Cos θ) (1) where λ is the wavelength of the incident light, β is the full width at half-maximum (FWHM) and θ is the diffraction angle. The FWHM value increased with the increase of TS from 300 oC to 400 oC and it decreases thereafter, indicating good crystallinity for the films formed at 400 oC. The evaluated average crystallite size varied in the range 28-54 nm with the increase of TS and is shown in Fig. 3. The increase of crystallite size with TS is attributed to the coalescence of many smaller nuclei to grown into bigger ones improving the crystallite size.
Fig. 3. Variation of crystallite size in SnS films as a function of sulphidisation temperature.
Figure 4 shows the scanning electron micrographs of Sn-precursor layers sulphidised at three different temperatures. The coalescence of smaller crystallites into bigger ones can be clearly seen when the temperature increased from 350 to 400 oC. The bright crystals on the surface were identified as SnS from the EDAX measurements; these were approximately stoichiometric in composition. The number of such bright crystals increased with the increase of sulphidisation temperature from 350 to 400 oC, indicating that such temperatures favour SnS formation. However, these crystals completely disappeared when the temperature was raised to 450 oC, probably due to the re-evaporation of S from SnS due to its high vapour pressure.
Fig. 4. SEM pictures of tin layers sulphidised at temperatures (a) 350 oC; (b) 400 oC and (c) 450 oC.
(a) (b) (c)
376 M. Leach et al. / Energy Procedia 15 (2012) 371 – 3786 M. Leach et al. / Energy Procedia 00 (2011) 000–000
Reflectance versus wavelength data was obtained for SnS layers rather than the transmittance versus wavelength data due to the Mo-coating at the back on glass substrates. Figure 5 shows the reflectance data plotted against wavelength for SnS layers sulphidised at three different temperatures. The spectra showed a pronounced rise of reflectance below the fundamental absorption edge corresponding to the energy bandgap of the layer. In present study, the reflectance results can be interpreted in this way because photons with energies less than the energy bandgap of the SnS will be transmitted through the tin sulphide layer to the Mo layer where they are reflected back out of the structure. On the other hand photons with energies greater than the energy bandgap will be absorbed by the SnS and therefore not transmitted or reflected. The reflectance data is used to calculate the optical absorption coefficient using the following relation.
α = -(1/d) ln (R(%)), (2) where is the absorption coefficient, R (%) is the reflectance taken from Fig. 5 and d is the thickness of the layer. In this work, the measured thicknesses of the individual layers, synthesised at different temperatures, were 193 nm, 200 nm and 208 nm. The value of d used for calculation of was taken as 200 nm, averaging the thicknesses of the individual films.
Fig. 5. Reflectance versus wavelength spectra of SnS films.
Using the absorption coefficient and photon energy, the energy bandgap of the films can be calculated using the relation [13]
αhν = A(hν - Eg)1/2 (3) where A is a constant and Eg is the energy bandgap. The intercept of the straight line drawn in (αhν)2 vs. hν plot on to the x-axis, as shown in Fig. 6, directly gives the energy bandgap of the material. The evaluated energy bandgap of SnS films varied between 1.31 and 1.36 eV, which is in good agreement with the reported values [14, 15]. The observed higher bandgap at lower TS is probably due to the presence of other phases that have higher bandgaps. With an increase of TS the energy bandgap decreases;
377M. Leach et al. / Energy Procedia 15 (2012) 371 – 378 M. Leach et al. / Energy Procedia 00 (2011) 000–000 7
this can be attributed to the secondary phases disappear leading to single phase SnS with the increase of TS.
Fig. 6. Plot of (hν)2 vs. photon energy (hν).
4. Conclusions
It is possible to produce SnS by sulphidising Sn-precursors in H2S in which case the layers were found to have the (111) plane as the preferred orientation. For annealing at temperatures < 400 ºC the layers contained secondary phases such as SnS2 and Sn2S3. For temperatures near to 400 ºC single phase SnS was produced, although if the temperature was too high the material re-evaporated from the substrate surface. The energy bandgap was found to be in the range, 1.31-1.36 eV, the higher value corresponding to presence of the secondary phases. The structural properties of the layers was similar to that produced by annealing in elemental sulphur; however the temperature needed to produce single phase SnS is lower in the latter case. Given that H2S is often used in very large scale industrial processes it may be a better method than elemental sulphur for producing large areas of SnS-based solar cells, despite the toxicity of H2S and the higher temperature needed for SnS synthesis.
Acknowledgements
The authors K.T.R. Reddy and R.W. Miles, Northumbria University and Sri Venkateswara University wish to acknowledge the support from the European Commission to conduct part of this work as part of the Marie Curie support scheme.
References
[1] Miles RW, Zoppi G, Reddy KTR, Forbes I. Thin film solar cells based on the use of polycrystalline thin film materials. In: Zhang S, editor. Organic nanostructured thin film devices and coatings for clean energy, London: CRC Press; 2010, p. 1-56.
378 M. Leach et al. / Energy Procedia 15 (2012) 371 – 3788 M. Leach et al. / Energy Procedia 00 (2011) 000–000
[2] Reddy KTR, Prathap P, Miles RW. Thin films of tin sulphide for application in photovoltaic solar cells. In: Tanaka H, Yamashita K, editors. Photovoltaics: developments, applications and impact, New York: Nova Science Publishers; 2010, p. 1-27.
[3] Gao C, Shen H, Sun L. Preparation and properties of zinc blende and orthorhombic SnS films by chemical bath deposition. Appl. Surf. Sci. 2011; 257: 6750-5.
[4] Kang F, Ishimura M. Pulsed electrodeposition of oxygen free tin monosulphide thin films using lactic acid/sodium lactate buffered electrolytes. Thin Solid Films 2010; 519: 725-8.
[5] Reddy NK, Reddy KTR. Preparation and characterization of sprayed tin sulphide films at different precursor concentrations. Mater. Chem. Phys. 2007; 102: 13-8.
[6] Ogah EO, Zoppi G, Forbes I, Miles RW. Thin films of tin sulphide for use in thin film solar cell devices. Thin Solid Films 2009; 517: 2485-8.
[7] Devika M, Reddy NK, Ramesh K, Ganesan V, Gopal ESR, Reddy KTR. Influence of substrate temperature on surface structure and electrical resistivity of the evaporated tin sulphide films. Appl. Surf. Sci. 2006; 253: 1673-6.
[8] Wiedemeier H, Csillag FJ. Equilibrium sublimation and thermodynamic properties of SnS. Thermochimica Acta 1979: 34: 257-65.
[9] Wiedemeier H, Csillag FJ, Gaur U, Wunderlich B. Heat capacity measurements on Sn2S3 and the thermodynamic functions of the tin sulfides. Thermochimica Acta 1980: 35: 187-96.
[10 Piacente V, Foglia S, Scardala P. Sublimation study of the tin sulphides SnS2, Sn2S3 and SnS. J. Alloy Compd. 1991; 177: 17-30.
[11] Devika M, Reddy KTR, Reddy NK, Ramesh K, Ganesan R, Gunasekar KR and Gopal ESR. Microstructure dependent physical properties of evaporated tin sulphide films. J. Appl. Phys. 2006; 100: 023518:1-7.
[12] Malaquias J, Fernandes PA, Salomé PMP, da Cunha AF. Assessment of the potential of tin sulphide thin films prepared by sulphurization of metallic precursors as cell absorbers. Thin Solid Films 2011: 519: 7416-20.
[13] Reddy KTR, Reddy PP. Structural studies on SnS films grown by a two-stage process, Mater. Letts. 2002: 56: 108-11. [14] Minemura T, Miyauchi K, Noguchi K, Ohtsuka K, Nakanishi H, Sugiyama M. Preparation of SnS by low temperature
sulfurization. physica status solidi c 2009: 6: 1221-4. [15] Warren BE. X-ray Diffraction. 2nd ed. New York, Dover; 1990. [16] Pankov JI. Optical Process in Semiconductors. 2nd ed. New York, Academic Press; 1975. [17] Reddy NK, Reddy KTR. Optical behaviour of sprayed tin sulphide thin films. Mater. Res. Bull. 2006; 41: 414-22. [18] Cheng S, Chen Y, He Y, Chen G. The structure and properties of SnS thin films prepared by pulsed electrodeposition. Mater.
![ALH [99] Gases Nobles.ppt](https://static.documents.pub/doc/80x56/577ce14a1a28ab9e78b52c6e/alh-99-gases-noblesppt.jpg)
